The mammalian genomes have purged themselves of CpGs because they constitute hotspots for mutations. Most of the remaining CpGs are methylated to protect aginst mutation susceptibility, but about one third of the remaining CpG dinucleotides have over time coalesced into patches referred to as CpG Islands (CGIS) that tend to be unmethylated an are used in biology for regulation of gene expression. We have developed a novel and cost-effective method to survey the methylOme in which we address, on a genome wide basis, one in three GpGs as to their metylation status, and one in two CpGs in CGI regions. In numbers, there are 21 million CpGs in the mouse genome (28 in humans) of which we address close to 7 million and are able to measure thir methylation status with single dinucleotise resolution. This method has just been published. An initial and important finding has been that classic definitions of CGIs, which are based on CpG density and constitute computer algorithms, instead of actual measurement, have failed in locting many unmethylated CpG patches - which we call non-CGI unmethylated regions (non-CGI UMRs). We found that these so-far unrecognized UMRs are significantly enriched in validated regulatory elements thtat can be found in the ORegAnno (open regulatory annotation) database. Inthis coming year we plan to develop the bioinfomatics tools that will allow us to define - if possible- tissue specific differences as well as treatment specific differences in the methylome. These may serve to diagnose diseases and changes due to environmental exposures.